G

When Bank of America (NYSE:BAC) purchased Merrill Lynch at the nadir of the financial crisis, the deal seemed too good to be true. For years, the Charlotte-based lender had coveted the synergies that promised to be unlocked by a marriage of Merrill Lynch's investment bank and thundering herd of wealth advisors to Bank of America's coast-to-coast network of retail branches.

But this lofty dream has since turned into a nightmare. While the lasting damage caused by the union has been shielded from sight by a mountain of legal issues, underneath these problems lies a universal bank that's spread too thin, outmatched by competitors, and should be broken up.

An uneven playing field: Part 1
The most obvious place this shows up is in Bank of America's net interest income, which, for all intents and purposes, is the top line item on its income statement. Keep in mind that commercial banks make most of their money through interest rate arbitrage; they borrow at low short-term interest rates and then lend the same funds out at higher long-term rates.

On the one hand, Bank of America has done a commendable job at this, as it has accumulated the largest deposit base in the United States -- deposits being the cheapest source of funding available. According to the FDIC, Bank of America has $1.17 trillion in deposits versus $1.08 trillion at JPMorgan Chase (NYSE:JPM) and $1.07 trillion at Wells Fargo (NYSE:WFC).

On the other hand, however, Bank of America comes up woefully short. You can see this in its yield on earning assets. In the third quarter of this year, it earned 2.87% on its average invested assets compared to 3.34% at Wells Fargo and 3.58% at U.S. Bancorp (NYSE:USB).

The problem is that Bank of America, due to its investment banking activities, must keep a larger share of its earning assets in low-yielding but highly liquid forms such as cash, government bonds, and mortgage-backed securities issued by Fannie Mae or Freddie Mac. Meanwhile, purer play commercial lenders are free to allocate a bigger share of assets to higher-yielding but less liquid loans.

G

The result is that Bank of America's net interest margin, which is a proxy for the profitability of its balance sheet, is destined to be lower than the commercial lenders it competes against. Just looking at its main banking subsidiary, for instance, Bank of America's net interest margin was 2.4% in the three months ended Sept. 30 compared to 3.1% at Wells Fargo and 3.4% at U.S. Bancorp.

An uneven playing field: Part 2
To be fair, a low net interest margin isn't unexpected from a firm that's split between commercial and investment banking operations. This is because investment banks primarily generate noninterest income, which comes from activities like trading, advising on mergers and acquisitions, and underwriting the issuance of debt and equity securities.

Take JPMorgan as an example. The net interest margin at its primary banking subsidiary was an anemic 1.98% last quarter. However, the nation's largest bank by assets more than made up for this with noninterest income, which accounted for the majority, or 54%, of its net revenue. On an annualized basis, that equates to 2.6% of earning assets versus 2.4% at Bank of America.

Now, if you pause for a moment and consider the proximity of the latter two numbers, you'd be excused for concluding that Bank of America isn't in fact very far off the mark in terms of noninterest income. JPMorgan is, after all, the nation's marquee financial institution. But what's important to keep in mind is that it costs JPMorgan less on a relative basis to generate more fee-based revenue.

At Bank of America's main subsidiary, personnel and occupancy costs equated to 1.64% of average assets last quarter. That was better than JPMorgan's 1.68%. However, once you include other overhead expenses and reduce the figure by noninterest income, then the relationship inverts by a large margin; Bank of America comes in at 0.97% versus JPMorgan's 0.60%. What this tells us is that JPMorgan gets much more noninterest income from its overhead expenses on a relative basis than Bank of America does.

G

Thus, in addition to the fact that Bank of America generates less net interest income than better-heeled commercial lenders like US Bancorp and Wells Fargo, it also costs the North Carolina-based bank more to produce less noninterest income than universal lenders with substantial investing banking operations such as JPMorgan. One could argue, in other words, that the combination has left Bank of America with more of the bad and less of the good from the disparate pieces.

And to make matters worse, Bank of America's investment banking activities have increased its operating costs -- personnel expenses, in particular -- relative to other commercial lenders. For example, despite the fact that Wells Fargo has 34,000 more employees at its primary subsidiary, the division's compensation ratio is actually less than that of Bank of America's analogous unit. This follows from the fact that the average pay at Bank of America's unit is $124,000 compared to Wells Fargo's $92,000.

You can attribute this gap to multiple things. Among others, it reflects that Wells Fargo has earned its reputation of operating efficiently, while Bank of America is, and essentially always has been, known for a bloated expense base that robs long-term shareholders of a respectable return on investment. And perhaps even more to the point, Bank of America's elevated compensation ratio (relative to employee count) likely stems from the fact that no self-respecting investment banker would be caught dead earning less than six figures.

Breaking up Bank of America
The net result is that Bank of America seems to have unlocked more bad synergies from the marriage of its vast retail branch network to Merrill Lynch's wealth management and investment banking activities than good ones. Namely, its balance sheet can no longer compete on a level playing field with first-rate commercial lenders like US Bancorp and Wells Fargo; and it will always be a second-rate player (and thus profiteer) on Wall Street compared to the likes of JPMorgan.

Does this mean Bank of America should be broken up? While investors will get a better sense for this as fewer legacy issues cloud the analysis in the years to come, the facts as we know them today suggest that, at the very least, a divorce shouldn't be taken off the table any time soon.

John Maxfield has no position in any stocks mentioned. The Motley Fool recommends Bank of America, Goldman Sachs, and Wells Fargo. The Motley Fool owns shares of Bank of America, JPMorgan Chase, and Wells Fargo. Try any of our Foolish newsletter services free for 30 days. We Fools may not all hold the same opinions, but we all believe that considering a diverse range of insights makes us better investors. The Motley Fool has a disclosure policy.